Hi ,
I am still unclear why in the Agentic AI course, Python code was required to build an Agent for example to reflect and review the SQL code it wrote in Module 2. Here the AI model is being called upon in the code, why not build all of it in the prompt itself by detailing out the steps in the prompt?
You could put the steps into a prompt, but the important difference is that an agent uses code to orchestrate actions and interact with external tools. In the SQL example, Python allows the system to generate SQL, execute it, capture the result or error, send that information back to the AI for review, and then act on the revised SQL. The prompt tells the AI what to do, while the code connects the AI to the tools and controls the workflow. Python is commonly used for this because of its strong AI/ML and automation ecosystem, although it is not the only language that can be used.